Name
[Windows Manager]
Installation_folder\SVROR\Manager\bin\rcxadm mgrctl - Manager control
[Linux Manager]
/opt/FJSVrcvmr/bin/rcxadm mgrctl - Manager control
Format
rcxadm mgrctl start rcxadm mgrctl stop rcxadm mgrctl modify {-ip ip |-port name=number} rcxadm mgrctl snap [-dir directory] [-full|-all]
Description
rcxadm mgrctl is the command used to start and stop managers, collect troubleshooting data, and change admin LAN IP addresses and port numbers.
For information on starting and stopping managers, refer to "2.1 Starting and Stopping Managers" in the "Operation Guide CE".
For information on collecting troubleshooting data, refer to "1.1.1 Collecting Initial Troubleshooting Data" and "1.1.2 Collecting Exhaustive Troubleshooting Data" in "Troubleshooting".
For information on methods for changing IP addresses and port numbers, refer to "6.1 Changing Admin IP Addresses" and "6.2 Changing Port Numbers" in the "User's Guide for Infrastructure Administrators (Resource Management) CE".
[Windows Manager]
If port numbers are changed, the following lines in the Windows_system_folder\system32\drivers\etc\services file are automatically updated.
[Linux Manager]
If the port number has been changed, the following lines in the \etc\services file will be changed automatically.
# service name port number/protocol name rcxmgr 23460/tcp rcxtask 23462/tcp rcxmongrel1 23463/tcp rcxmongrel2 23464/tcp rcxmongrel3 23466/tcp rcxmongrel4 23467/tcp rcxmongrel5 23468/tcp rcxdb 23465/tcp nfdomain 23457/tcp nfagent 23458/tcp |
Note
In a clustered manager configuration, this command should not be used to start or stop the manager.
[Windows Manager]
Start the Manager
In the Failover Cluster Management tree, right-click [Services and Applications] for managers, and select [Bring this service or application online] from the popup menu.
Stop the Manager
In the Failover Cluster Management tree, right-click [Services and Applications] for managers, and select [Take this service or application offline] from the popup menu.
[Linux Manager]
Use the cluster system administration view (Cluster Admin) to start or stop the manager.
Refer to the PRIMECLUSTER manual for details.
For Basic mode, the following values are not written.
rcxmongrel3
rcxmongrel4
rcxmongrel5
Subcommands
Starts the manager.
Stops the manager.
Modifies the IP address used by the manager in the admin LAN or its port number.
Collects troubleshooting data.
Options
Specify the following options when using the modify subcommand:
Specify the new manager IP address.
Specify the port name to be modified in name, and the new port number in number.
The port name can be one of the following values:
rcxmgr
rcxtask
rcxmongrel1
rcxmongrel2
rcxmongrel3
rcxmongrel4
rcxmongrel5
rcxdb
nfdomain
nfagent
Note
For Basic mode, the following values cannot be specified.
rcxmongrel3
rcxmongrel4
rcxmongrel5
The following options can be specified for the snap subcommand:
Specify the folder used to store the collected data in directory.
When omitted, the data will be stored in the folder specified by the TEMP environment variable:
Note
When using full paths in the dir and the TEMP environment variable
The length of the full path string must not exceed 100 characters. If more than 100 characters are used the troubleshooting data cannot be collected, and message number 67131 or message number 67265 will be displayed.
When using relative paths in the dir and the TEMP environment variable
When specifying a relative folder path, its equivalent full path must not exceed 100 characters (calculated using the Windows 8.3 format (*)). If the converted full path string exceeds 100 characters, the troubleshooting data will not be collected, and the "Message number 67131" will be displayed.
* Note: This rule specifies that the file name can be a maximum of 8 characters, with a file extension of up to 3 characters
The following symbols cannot be specified in the name of the folder in which the collected data is stored:
""", "|", "*", "?", "/", "<", ">", ",", "%", "&", "^", "=", "!", ";"
[Windows Manager]
"/"
[Linux Manager]
"\"
Collects exhaustive troubleshooting data from the admin server. This data is required to isolate the cause of a problem which could not be identified from initial troubleshooting data alone.
This requires significantly more disk space for the generated data files. This option can be omitted when collecting troubleshooting data for an initial investigation (first diagnostic).
This option collects troubleshooting data not only from the admin server, but from all the managed servers as well. This option cannot be used together with the -full option.
Since only data required for initial troubleshooting is collected, it does not require much disk space. The data can also be collected quickly and sent easily via email.
Data collected from both the admin server and managed servers is stored in the directory specified by dir on the admin server from which the command was executed.
If the user account does not have administrative authority within the operating system, it is necessary to log in as a Resource Orchestrator privileged user (using the rcxlogin command) to use this option.
For information on the rcxlogin command, refer to "2.1 rcxlogin".
Collected Files
The collected data is stored in the following compressed files:
Admin Server
[Windows Manager]
rcxtssnap_server_name.jar
The server_name part will be in lower case when the -all option is specified, or upper case when omitted.
[Linux Manager]
rcxtssnap_server_name.tar.bz2
The server_name part will be in lower case when the -all option is specified, or upper case when omitted.
Managed Server
[Windows] [Hyper-V]
Managed_server\rcxtssnap_physical_server_name.jar
[Linux] [VMware]
Managed_server\rcxtssnap_physical_server_name.tar.bz2
The managed server's name is displayed in Managed_server.
Note
[Linux] [VMware]
When collecting troubleshooting data, data is compressed on managed servers using either the bzip2 or the gzip compression command. Depending on the command used, the resulting file extension will be either one of the following.
Resource Orchestrator uses the command with the best compression ratio (bzip2 -> gzip) available on the managed server.
- When compressing with bzip2
*.tar.bz2
- When compressing with gzip
*.tar.gz
Execution Log
Data Collection Results
rcxtssnap_result.txt
This file displays collection results in the following format:
Server_name:Result |
Server_name
For a managed server, the physical server name is displayed.
For an admin server, the server name is displayed.
Result
OK: Indicates that the data collection was successful.
NG: Indicates that the data collection failed.
Example
blade1:OK |
Error Log
Admin server
rcxtssnap_server_name_error.txt
Managed server
[Windows]
Managed_Server\rcxtssnap_physical_server_name_error.txt
[Linux]
Managed_Server\rcxtssnap_physical_server_name_error.txt
Requirements
OS Administrator
Note that when logged in as a Resource Orchestrator privileged user (using the rcxlogin command), only the -all option of the troubleshooting data collection subcommand (snap) can be used.
Admin server
Exit Status
This command returns the following values:
The command executed successfully.
An error has occurred.
